| China as one of the largest coal production and consumption country around the world, it is also one of the accident occurred frequently country such as fire damp explosion, water burst, catch on fire during the exploitation of coal mine. The research of detecting by coal mine shaft underground explorative robot make the coal mine rescue works implements smoothly, to reduce casualties caused by mine disaster, it will be substituted for or partial substituted for emergency workers to enter the mine shaft disaster site and to carry out environmental prospecting and surveying task. The topic is from the scientific and technological project in Shanxi Province, " mine underground rescue detection robot technology and system research ", this thesis is mainly research the exploration robot motion control that related issues, including the detection and modeling of the external environment;the motion obstacle-climbing capability and motion control research, path planning and other issues. Underground mine detection robot motion control issues related to the successful completion of the research is the key to the detection task.Harsh environment for the coal mine, there are methane, carbon monoxide and other combustible gases, poor road conditions. Studied the detection of the robot control system, it uses embedded computer with PC and lower place machine DSP and CPLD combined. Circuit system fully adopts the design of Ann circuit design, and flame retardant gas explosion protection design. The robot uses sonar sensors, PSD sensor and video capture features of the combination of multi-sensor data fusion layer to detect the external environment. First data after detecting to extremum average filtering, then according to the characteristics of each sensor to make the credibility of fusion, and through experiment testing this data processing method is accurate data and possesses very strong anti-interference ability.According to the structural characteristics of coal mine detection robots, I deduced the characteristics of the kinematics equations and the speed of the two crawler formula when the robot is in linear motion and turns a corner, and proposes making use of fuzzy self-tuning PID control system to control the robot's low-level motor sport, and using Simulink software to compare with the simulation the control system, PID control systems and fuzzy control system. Meanwhile analysis detection robot obstacle-navigation method and ability, the use of MATLAB simulation detected when the robot obstacle-navigation changes and applied it to the center of mass of actual obstacle-navigation.The main task of underground mine detection robot path planning is in the aftermath of the downhole obstacles uncertain environment, from a starting point to target path, looking for the safety of avoidance path which is shorter and consume the minimum energy. Combined with the characteristics of coal mine underground detection robots, if it is able to span the tested obstacle, it will span; if the obstacle is unable to span, It will use a round of "window scanning scrolling tra- reaction" hybrid control the path planning. Round rolling window is divided into three regions:(1) when the emergency obstacle-avoiding areas appear obstacle, robot response promptly and transform movement direction to prevent robots collisions. (2) when local figure planning area appear obstacles, path planning divide kind judgment the movement direction of designing robots, then make changes ahead of anticipation to change direction makes detection robot can through along the edge of obstruction within the shortest distance. (3) When fuzzy planning area has blocking, as long as follow the target set into motion, the movement direction of robot can get through it. While the measurement data of sensors is probably not very accurate, therefore the detecting robot needs approached again, then makes accurate map planning. Finally, the path planning method by MATLAB simulation verifies the effectiveness of the algorithm. |
PDF Full Text Request